OpenRR-1k is a 1,000-pair real-world reflection removal dataset generated by off-the-shelf AI removal plus manual refinement, with benchmarks showing fine-tuning gains.
Our pipeline provides researchers with a more convenient way to collect diverse, high-quality true real- world data samples at low cost
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
OpenRR-1k: A Scalable Dataset for Real-World Reflection Removal
OpenRR-1k is a 1,000-pair real-world reflection removal dataset generated by off-the-shelf AI removal plus manual refinement, with benchmarks showing fine-tuning gains.